Research in this laboratory is directed to investigations of mechanisms whereby compounds which exhibit mutagenicity in short-term assays such as the Ames/Salmonella test may fail to induce carcinogenesis in rodents under conditions of the NTP 2-year bioassay. this nonconcordance between short- term tests and NTP bioassay results weakens the use of in vitro assays for prediction of in vivo carcinogenicity. The results of this research indicate that the mutagenic noncarcinogens consistently fail to induce cell proliferation in vivo. Similar work in this laboratory with positive carcinogens, either mutagenic or nonmutagenic, demonstrated each of these chemicals did induce cell proliferation, and further, the cell proliferation was associated with the site of carcinogenesis. We have studied 17 chemicals to date (15 in-house; 2 on contract) which vary considerably in their structural classes. A series of 5 organophosphate compounds demonstrated that the potent renal carcinogen tris (2,3- dibromopropyl)phosphate produces cell proliferation in the renal outer medulla, the site of renal tumors in the bioassay. Noncarcinogens in this class did not produce cell proliferation in liver or kidney. A 90 day feeding study was conducted in-house on the anxiolytic agent Oxazepam which combined cell proliferation measurements with clinical chemistry and pharmacokinetics. Our data demonstrated that this compound may be carcinogenic in B6C3F1 mice due to its ability to produce cell proliferation by a mitogenic mechanism. The results of the research conducted in this laboratory suggest a positive correlation between cell proliferation and carcinogenesis for these chemicals, irrespective of their mutagenicity.